Peter Ellis’ convictions, while long ago, were not final. Unfortunately, neither will be the outcome of his appeal to the Supreme Court.

Ellis had been granted an extension of time to appeal but died before the appeal could be heard. Therefore, a decision had to be made on whether to allow the appeal to proceed, which meant deciding whether the issues in the appeal had become moot. Before examining that important procedural decision, I first summarise the outcome of the successful appeal against conviction.

Before the Supreme Court, questions were raised about the reliability of the complainants’ evidence, as child witnesses, particularly regarding suggestibility, contamination and unsound expert evidence about their observed behaviours.

The court ruled on these issues in its substantive judgment (Ellis v R [2022] NZSC 115), allowing the appeal on the basis that there had been a miscarriage of justice. That holding amounted to the court determining that the combined unsatisfactory aspects of the case, particularly the evidence given by an expert under the now-repealed s 23G of the Evidence Act 1908, meant there was a real possibility of different verdicts.

Evidential problems 

Section 23G enabled qualified experts to give evidence about exhibited behaviours that were “…consistent or inconsistent with the behaviour of sexually abused children of the same age group as the complainant.”

The court held that the Crown expert’s evidence on the behaviours of the children was not balanced; her evidence also failed to adhere to acceptable standards. The expert did not inform the jury of other possible causes of the children’s behaviours and, when she did so, she discounted or minimised other possible causes. Her evidence also involved circular reasoning.

The Supreme Court observed that a strict approach had been taken to the rules governing admission of s 23G evidence, with departure from them resulting in some cases in a miscarriage of justice, citing the 1991 decision in Crime Appeal (CA244/91) v R, where Hardie Boys J, in delivering the judgment of the Court of Appeal, said the limits of evidence given under the section were so clearly in place that any exceeding of them must be regarded as serious, because “[o] ne cannot know what influence it will have on a jury”. It follows that by the date of Ellis’ trial, the risks associated with such evidence were well-established.

The court also concluded that while the risk of contamination of the children’s evidence was canvassed at the trial, the jury was not adequately informed about the level of risk of contamination. Further, incomplete evidential records went before the jury, and medical evidence (from an examination of some of the children) was flawed (this is a class of evidence that was criticised by the Court of Appeal in 2008, see R v Garraway [2008] NZCA 2). The Supreme Court did not hold that Ellis was probably innocent or that the evidence was insufficient to sustain a conviction. Rather, it held that the trial had gone badly wrong.

Ordinarily, had the appeal been allowed closer to the trial, there would have been a retrial. But, as the Supreme Court observed, Ellis did not promptly exercise his right to seek permission to bring an appeal before the Privy Council (the court’s judgment did not convey why Ellis had not done so); nor had he promptly sought the Crown’s agreement to bring an appeal before the Supreme Court – ie, after the court had been constituted in 2003 (the Crown’s later agreement for him to do so allowed the Supreme Court to hear the appeal). While all the judges agreed that the court had the power to hear the appeal despite Ellis’ death, had the minority judges controlled this gateway decision the appeal would not have been heard.

Tikanga Māori 

All the judges explored tikanga. There was broad agreement about the principles of tikanga since they had been agreed after a wananga of the parties’ lawyers and experts. But this was not a case where tikanga arose as part of the context or subject matter of the underlying litigation. In fact, the question arose only after it was raised by a member of the Supreme Court bench, after it became apparent that, if the appeal was to be heard, it would have to be heard posthumously.

The minority dissenting judgment of Justices Arnold and O’Regan is my focus in the balance of this article (I am sure the majority’s reasoning will be fulsomely discussed by other commentators). The minority judges opined that this was not a suitable case for the court to make pronouncements of a general nature about the incorporation or application of tikanga as part of New Zealand’s common law.

In my opinion, their dissent was for orthodox reasons, having little to do with tikanga per se. While respecting the work done on the issue, the minority observed that because tikanga had come up for consideration only in the context of an interlocutory matter (before the Supreme Court), it was not appropriate for the court to extensively explore and opine upon it. The court did not have the benefit of decisions from lower courts; the issue arose in “an uncontested environment” and the court had heard no contrary argument.

The minority pointed to a number of difficult issues of legal and constitutional significance that might arise, including: how the court can identify when tikanga is relevant to the case at hand and when it is not; if it is relevant, how it should be addressed; whether tikanga is a separate or third source of law; how the relevant tikanga should be brought to the court’s attention (noting the acknowledgement in the reasons of Glazebrook J that the process used in this case, though commendably thorough and authoritative, cannot be followed in more run-of-the-mill cases); how the application of tikanga in one area of the law affects the common law in another area; and how to avoid tikanga being distorted when applied by courts (a concern shared by Williams J).

The minority thought tikanga did not appear to have materially affected the opinion of Glazebrook J about whether to give permission to continue the appeal. The implication is that the majority did not need to rely on tikanga when deciding whether to give permission to hear the appeal posthumously (if that is what they thought was the just decision on the application of ordinary principles).

The minority added that incorporating tikanga Māori values in a criminal process, which usually proceeded on the basis of a very different set of values, was not straightforward. There is force in these arguments.

Despite the extensive publicity over the years about the safety of Ellis’ convictions, the exploration and application of tikanga by the Supreme Court is what the Ellis appeal will be remembered for. The not-uncommon reasons for the court holding that there had been a miscarriage of justice are likely to fade over time, as against the power of the first judgment as a precedent. The examination of tikanga in the context of a criminal appeal was not the ideal vehicle for the Supreme Court to explore the positioning of tikanga in the law of New Zealand.

As the minority observed, judicial pronouncements about the place of tikanga in a broader legal context might be better made following the work of the Law Commission (see Te Aka Matua o te Ture | Law Commission Tikanga Māori www.lawcom.govt.nz; as referenced by Glazebrook J, in footnote 138 of Ellis v R [2022] NZSC 114: “the study paper plans to explain tikanga Māori, its sources and its expression in the courts and the Waitangi Tribunal, with the aim of providing a framework for engagement with tikanga Māori within Aotearoa New Zealand’s legal system”).

Moreover, despite the aspirational aspects of tikanga for the resolution of conflict in this context, the outcome of the appeal has nonetheless left open the question of Ellis’ guilt for serious sexual offending.

While it should not be overlooked that Ellis served a long prison sentence, there is no finality for the complainants or for Ellis’ family. The complainants and Ellis’ family do not have the option of final verdicts after a fair trial. Given the nature of the questions raised in the appeal and observing that there was no ground of appeal upheld that the verdicts were unreasonable, this outcome was entirely predictable, even on a best-case scenario for a posthumous appeal in this case. The Ellis case is more proof of the maxim that justice delayed is justice denied, for all involved. ■

Warren Pyke is an Auckland barrister ■